An integrated manifold learning approach for high-dimensional data feature extractions and its applications to online process monitoring of additive manufacturing

Abstract

As an effective dimension reduction and feature extraction technique, manifold learning has been successfully applied to high-dimensional data analysis. With the rapid development of sensor technology, a large amount of high-dimensional data such as image streams can be easily available. Thus, a promising application of manifold learning is in the field of sensor signal analysis, particular for the applications of online process monitoring and control using high-dimensional data. The objective of this study is to develop a manifold learning-based feature extraction method for process monitoring of Additive Manufacturing (AM) using online sensor data. Due to the non-parametric nature of most existing manifold learning methods, their performance in terms of computational efficiency, as well as noise resistance has yet to be improved. To address this issue, this study proposes an integrated manifold learning approach termed multi-kernel metric learning embedded isometric feature mapping (MKML-ISOMAP) for dimension reduction and feature extraction of online high-dimensional sensor data such as images. Based on the extracted features with the utilization of supervised classification and regression methods, an online process monitoring methodology for AM is implemented to identify the actual process quality status. In the numerical simulation and real-world case studies, the proposed method demonstrates excellent performance in both prediction accuracy and computational efficiency.

Keywords:

• Additive manufacturing
• integrated manifold learning
• isometric feature mapping (ISOMAP)
• multi-kernel metric learning (MKML)
• online process monitoring

Additional information

Funding

The research reported in this publication was supported by the National Science Foundation under Award Number CMMI 1436592 and the Office of Naval Research under Award Number N00014-18-1-2794. Part of the research supported from the U.S. Department of Energy, Office of Energy Efficiency and Renewable Energy, Advanced Manufacturing Office, under contract DE-AC05-00OR22725 with UT- Battelle, LLC. This manuscript has been authored by UT-Battelle, LLC under Contract No. DE-AC05- 00OR22725 with the U.S. Department of Energy.

Notes on contributors

Chenang Liu

Chenang Liu received his double BS degrees in environmental & resource sciences and mathematics from Zhejiang University, China, in 2014; he then earned his MS degree in statistics and PhD degree in industrial and systems engineering from Virginia Tech in 2017 and 2019, respectively. He is currently an assistant professor in the School of Industrial Engineering and Management at Oklahoma State University. His research interests include data-driven analytics, process monitoring and control, and machine learning techniques for smart manufacturing and healthcare applications.

Zhenyu (James) Kong

Zhenyu (James) Kong received his BS and MS degrees in mechanical engineering from Harbin Institute of Technology, Harbin, China, in 1993 and 1995, respectively, and his PhD degree from the Department of Industrial and System Engineering, University of Wisconsin–Madison, Madison, WI, USA, in 2004. He is currently a professor with the Grado Department of Industrial and Systems Engineering, Virginia Tech, Blacksburg, VA, USA. His research interests include sensing and analytics for smart manufacturing, and modeling, synthesis, and diagnosis for large and complex manufacturing systems.

Suresh Babu

Sudarsanam Suresh Babu received his B. Engg., degree in metallurgical engineering from P. S. G. College of Technology, in 1986 Coimbatore, India, M. Tech degree in industrial metallurgy – welding from Indian Institute of Technology (Madras), Chennai, India in 1988 and his Ph.D. from University of Cambridge, UK in 1992. He is currently, the UT-ORNL Governors’ chair professor of advanced manufacturing in the Department of Mechanical, Aerospace and Biomedical Engineering and also part of the Energy and Transportations Sciences Division. He also serves as director of the Bredesen Center for Interdisciplinary Research and Graduate Education.

Chase Joslin

Chase Joslin received his BS degrees in materials science and engineering and in mathematics from the University of Tennessee, Knoxville in 2017. He began working at the Manufacturing Demonstration Facility (MDF) with Oak Ridge National Laboratory (ORNL) in 2018 and is currently a Technical Professional with the Deposition Science and Technology group. His research support includes operation, parameter development, and scan strategy optimization for electron beam powder bed fusion and laser powder bed fusion.

James Ferguson

James Ferguson received his BS in computer science from West Virginia University and his MS degree in computer science from University of Tennessee in 2014 and 2016, respectively. He worked at the Manufacturing Demonstration Facility (MDF) with Oak Ridge National Laboratory (ORNL) in the Imaging, Signals, and Machine Learning (ISML) group from 2016-2019. While at ORNL his main research areas were data analysis, defect detection, and scan strategy optimization for electron beam powder bed fusion. He currently works as a Software Development Engineer for Microsoft.